Just where exactly are the threatened species?
A georeferenced database of threatened species becomes imperative as human activities encroach more and more on fragile ecosystems, say Lawrence T. Ramos, Alfie M Torres, Florencia B. Pulhin and Rodel Lasco of the Forestry Development Center at the University of the Philippines at Los Baños and the World Agroforestry Centre Philippine Office.
Georeferencing, or putting on a map, the exact locations of where species occur is a prerequisite for many branches of science, such as modelling the distribution of species and analysing the correlation of species with ecosystems. Georeferencing also helps to track plant species and prioritize scarce resources for conservation.
Because of the importance of georeferencing and also the shortage of its deployment (more than 90% of the billion or more occurrence records found in biological specimens worldwide are not georeferenced), we developed a method to georeference occurrences of threatened forest tree species in the Philippines.
The Philippines is a tropical country hosting a high concentration of diverse plant species, ranking fifth in the world, and housing 5% of the world’s flora. Yet, ironically, it is also a leading global biodiversity hotspot of threatened forest trees owing to anthropogenic habitat alteration.
The Department of Environment and Natural Resources’ administrative order DAO 2007-01, which constitutes the official country listing of threatened plants, lists 174 vulnerable species, 101 critically endangered, 187 endangered and 64 other threatened species in the Philippines. The International Union for Conservation of Nature’s Red List of threatened plant species also provides an annually updated listing. Forest tree species are particularly threatened mainly because of demand for their timber.
To help address this, we used locality descriptions that were initially narrowed down in a geographic information system using administrative maps and then further confined using two criteria: 1) elevation; and 2) surface-cover information from remotely sensed images.
The result was a georeferenced database of 2067 occurrence records of 47 threatened forest species on a national scale, each with a unique location per species and enough metadata directing the database user to the source of occurrence data, based on locality descriptions, established species’ elevation range requirements, recent surface cover, and administrative and protected area boundaries.
The database can be used as a tool in determining priority species for collection of specimens or germplasm, for taxonomic identification and historical mapping. It also serves as an integral component in spatially modelling the distribution of tree species and forest formations in the past and in possible future scenarios.
Finding the information
Species’ occurrence information was taken from Global Biodiversity Information Facility entries and ‘legacy’, or historic, literature, primarily. There was much information contained in this legacy literature as well as in herbaria collections, both domestic and abroad. However, in the case of the Philippines, there were only a few, recent records in the Global Biodiversity Information Facility that had specific, fine-scale location information.
Species’ occurrence data before Global Positioning System (GPS) devices only mention localities, usually at the scale of provinces and municipalities. To compound the problem, names or boundaries of some localities have changed over the last hundred years. Formerly large provinces and districts have been subdivided, giving way to the creation of new government units.
Various authors have already contributed to documenting the occurrence of threatened forest tree species on a national scale. There are also numerous peer-reviewed journal articles, technical reports and biodiversity surveys done at watershed or protected-area scale that are good sources of occurrence data. But these are often analogue in form. If there happens to be georeferenced information, as is the usual practice today, the location data often stays with the collector and is not indicated in the tag on the specimen.
Overall, a great challenge now is the process of translating such sources from analogue to digital form to enable access by a wider public. In the case of legacy literature, digitization has made some headway already under the Biodiversity Heritage Library portal. For instance, the voluminous publications of Merrill, which document the former diversity and spatial distribution of Philippine forests are already digitized and can be accessed through the portal. Among Merrill’s significant publications is the four-volume Enumeration of Philippine Flowering Plants (1923–1926), which is the most comprehensive species’ list on a national scale to date.
Extensive work needs to be done, however, in the case of herbaria specimen collections, which remain largely analogue in form, if not entirely raw specimen tags. The discrepancy is easily appreciated if they are to be compared with biological information at the molecular and ecosystem levels, which are largely digitized. Another issue is access, for although biodiversity is concentrated in the developing countries, the wealth of scientific information in digitized form is concentrated in the libraries and natural history institutions of developed countries, probably very remote from the specimen’s origin.
Specimens contain basic, yet important information, for example, scientific name, collector’s name, collection date, and locality description, at the least. Traditionally, locality descriptions are based on names of places or situational landmarks that change over time. At best, a species’ locality description should be specific enough, leaving no room for uncertainty in interpretation.
In compliance with the Convention of Biological Diversity, the Philippines is an active participant in the ASEAN Clearing House Mechanism. The Mechanism is envisioned as a harmonized regional gateway of publicly available biodiversity information held by Association of Southeast Asian (ASEAN) member countries. However, the Mechanism is still in its infancy and is not yet included in the Global Biodiversity Information Facility network. There is yet to be a holistic, georeferenced and validated biodiversity information system on a national scale in the Philippines that aids decision-making and knowledge sharing.
A database of georeferenced species’ occurrence paves the way for visualization and higher forms of analysis. However, georeferencing can be tedious and suffers from major limitations: 1) it is slow; 2) the accuracy and precision of assigned coordinates are usually unknown; 3) a large fraction of available coordinates are inconsistent with locality information; 4) materials and methods are poorly documented; and 5) many localities are georeferenced many times over with different results.
The development of georeferencing solutions which are automated, interoperable and process-documented has long been sought by the scientific communities that rely on biological collections. The same sentiment has given rise to various projects that aim to provide tools for batch processing of locality descriptions and protocols for georeferencing biodiversity information.
Our work has only contributed to a small subset of solutions to these challenges. An effort like this on a national scale will probably take decades or so to fully complete. But the need to come up with a georeferenced database of threatened species’ occurrence becomes more imperative as human activities encroach more and more on fragile ecosystems, which are fast disappearing.
Edited by Robert Finlayson
Read the article
This work is related to the CGIAR Research Program on Forests, Trees and Agroforests